Katz and Ferretti's, (Discourse Processes, 2003, 36, 19) pioneering paper was the first to address and systematically examine the role of marking (literally speaking, in a manner of speaking, proverbially speaking) during online processing of proverbs (see also Schwint et al., 28th Annual Conference of the Cognitive Science Society Proceedings, 2006, 768). For Katz and colleagues, such markers function as introductory formulae, signaling to the addressee the intended interpretation of an incoming proverb. Inspired by their work, this paper explores the effects of marking, showing that some markers (literally, in the full sense of the word, double entendre, really) rather than disambiguating an ambiguous utterance, can allow for ambiguation (e.g., S/he is radiant, in the full sense of the word uttered in reference to a smiling person wearing sparkling clothes). Two offline questionnaire studies and one online reading task experiment, all conducted in Hebrew, test the Low-Salience Marking Hypothesis (Givoni, Low-salience marking, 2011; Givoni, Marking multiple meanings, 2020; Givoni, Journal of Pragmatics, 2013, 48, 29). Accordingly, such marking boosts low-salience meanings ("glittery", here the literal meaning) which are less-frequent, less-familiar, less-prototypical, and less-conventionalized (The Graded Salience Hypothesis, see Giora, Cognitive Linguistics, 1997, 8, 183; Giora, On our mind: Salience, context and figurative language, Oxford University Press, 2003; Givoni & Giora, Handbuch Pragmatik, J.B. Metzler, 2018). Marked utterances were embedded in contexts, strongly supportive of the salient meaning of the ambiguities ("happy", here the figurative meaning). Results support the Low-Salience Marking Hypothesis. They show preference for low-salience meanings as well as faster reading times of such meanings following low-salience marking relative to control conditions. (PsycInfo Database Record (c) 2021 APA, all rights reserved).